'Overlap' syndromes.

Many of us at the end of a weekly lupus clinic are tempted to reflect that almost all shades of connective tissue 'overlap' occur, and that it is perhaps pointless to attempt too precise a definition in individual cases. The introduction of the concept of overlap syndromes was a pragmatic way to resolve an imprecise compromise. It represented an attempt to categorise various groups of disorders having in common some clinical and serological features, but failing to fulfil the established or proposed criteria for the classification of a distinct and separate entity, such as rheumatoid arthritis, progressive systemic sclerosis, systemic lupus erythematosus, or dermatomyositis. Some claim that as many as 25% of patients with connective tissue disease fall into one or other overlap group. 1-5 In the past few years there has been limited success with attempts to describe distinct clinical entities, generally using serological markers. These include polemic entities such as mixed connective tissue disease,6'0 subacute cutaneous lupus erythematosus," 12 or the recently described 'primary' antiphospholipid syndrome, 1--6 among others. Following the observation by Sharp et al6 that an overlap syndrome consisting of Raynaud's phenomenon, swollen hands, arthritis, and myositis was strongly associated with the presence of antibodies to nuclear ribonucleoprotein (RNP), the concept of mixed connective tissue disease has been useful. The serological specificity of these antibodies has been defined to epitopes on a 68 kD phosphoprotein uniquely associated with the RNP containing uridylic acid-rich small nuclear RNA (U-snRNP). At least 12 U-snRNPs have been identified, which account for about 1% of the total cellular RNA. The Ul to U6-snRNPs, with the exception of U3-snRNP, are known to play an important part in messenger RNA processing. In addition, a number of core DNAs coding for these antigens have been cloned and some specific characteristics of the deduced amino acid sequences are known.'7 As the years go by clinical perception of mixed connective tissue disease has evolved considerably. For instance, patients with this disorder may have many other overlap and non-specific features, including pulmonary disease,'2' Sjogren's syndrome,22 and even, renal23 and cerebral disease.24 Therefore, diagnosis of this entity must be based both on clinical and serological criteria. Although some attempts at classification have been made,2527 no definite agreement has been achieved.28 29 In addition, it has been seen that the pattern of this disease often changes to one of progressive systemic sclerosis. Interestingly, in such patients, it is common to see a gradual reduction of the hyperglobulinaemia and disappearance of the antibodies to U-snRNP.' Finally, there have been no major advances in identifying the aetiopathogenesis of this condition or in explaining the apparent association with antibodies to U-snRNP.29 This has led to a protracted debate about the specificity or otherwise of mixed connective tissue disease.24 The title subacute cutaneous lupus erythematosus has been given to an overlap syndrome whose main clinical feature is the appearance of florid cutaneous lesions, often extremely photosensitive, sometimes with a characteristic serpiginous border. This disorder may include other systemic features such as Sjogren's syndrome, pericarditis, arthritis, and the occasional case of congenital heart block in the offspring (neonatal lupus). The serological association is with antibodies to SSA/Ro." 12 Recent knowledge of the structures that bear the SSA/Ro antigens has enabled definition of biochemical differences among patients with these antibodies. The first major antigen component of SSA/Ro to be described and cloned was a polypeptide of 6OkD. Immunoblot techniques have been used recently to identify two other polypeptides as 52 and 48 kD species. Antibodies against the newly recognised 52 kD SSA/Ro peptide component have been found to be predominant in the sera from mothers of infants with neonatal lupus, whereas antibodies against the 60 kD antigens were not related.30 The term antiphospholipid syndrome has been recently proposed for a group of patients who have antiphospholipid antibodies-for example, anticardiolipin antibodies or the lupus anticoagulant, and in whom multiple venous and arterial thrombosis, recurrent fetal loss, and thrombocytopenia are common. These patients may also develop other features, including heart valve lesions, livedo reticularis, chorea, and haemolytic anaemia.31-33 Whereas early studies concentrated on systemic lupus erythematosus or 'lupus-like' diseases,32 it is now recognised that these manifestations may occur without any other feature of lupus being present, leading to the definition of the primary antiphospholipid syndrome.13-16 Nevertheless, long term follow up of these patients has shown that some of them develop other connective tissue disease, such as systemic lupus erythematosus.34 On the other hand, it is surprising that patients with high titres of antiphospholipid antibodies do not necessarily develop any of these clinical manifestations. Lupus Research Unit, The Rayne Institute, St Thomas's Hospital, London SEI 7EH, UK R Cervera M A Khamashta G R V Hughes